CPU TASK Valid result
Evolving Personal Memory
GPT-6 Luna · Codex · reasoning effort max
Task score (%)32.8
Input Tokens15,087,765including 14,836,036 cached
Output Tokens191,631
Agent time45min
Cost$0.28
Recorded steps103
Cost is estimated from recorded agent tokens at official API rates in USD (2026-11-01; DeepSeek off-peak pricing).
Loading trajectory…
Select this tab to browse the saved files.
Coding time & token usage
- Original agent completion
- normal
- Coding started
- 2026-10-24T23:07:32.800426Z
- Coding finished
- 2026-10-24T23:52:43.811008Z
- Input (includes cache)
- 15,087,765
- Cached input (subset)
- 14,836,036
- Cache writes (subset)
- 246,986
- Output
- 191,631
- Reasoning (reported separately; not added)
- 136,197
- Recorded responses
- 100
Input includes cached input; output is counted once. Auxiliary reasoning counts are shown as recorded. Regrading does not repeat coding usage.
Select this tab to browse the saved files.
Configuration
Evaluation settings summary
{
"agent": "Codex",
"agent_timeout_sec": 7200,
"coding_commit": null,
"effort": "max",
"model": "openai/gpt-6-luna",
"provider": "OpenRouter",
"resources": {
"cpus": 8,
"gpus": 0,
"memory_mb": 32768,
"storage_mb": 102400
},
"run_mode": "official_harbor_full",
"task_commit": "bb34e915be5c2b2948146c2d7862cb58652d6300",
"verifier_commit": "bc2ef55cf5bc4e5133c32816a7874368192a2f1e",
"version": "0.164.0"
}Select this tab to browse the saved files.
Resolved configuration
Select this tab to browse the saved files.
Trial log
Select this tab to browse the saved files.
Exceptions
Select this tab to browse the saved files.
Select this tab to browse the saved files.
Trajectory downloads & provenance
Public text redacts private values and normalizes host paths. Source hashes identify the original records.
- Trial ID
evolving-personal-memory-gpt-6-luna-max-5f053fbb3ab8- Source trajectory SHA-256
a0c2a842f0a274d07b85e98fa49e43f32c44a969b66461d5421bbdc2ba4f6378- Final reward SHA-256
75467739e5bc69011854cea4b6c2277dbff203993f74e868de869b05cbd26976